The overall goal of this project is to identify the neural correlates of recovery from aphasia during the first three months after acute stroke. Aphasia is one of the most common and debilitating consequences of strokes affecting the dominant hemisphere. In the acute and subacute stages, most patients experience some degree of recovery of language function. These first few months after brain damage occurs are critically important because the greatest behavioral gains take place during this time, even when patients do not receive treatment. However the extent of recovery is highly variable: some patients recover most or nearly all language function, while others make few gains and remain chronically aphasic. It is not well understood what neural mechanisms underlie recovery during this period, nor why some patients recover so much better than others. To address these questions, we propose to study individuals with aphasia due to acute left hemisphere stroke. Patients will be examined at four time points after stroke: 24-48 hours, 2 weeks, 1 month, and 3 months. At each time point, a comprehensive yet time-efficient language evaluation will be administered to quantify expressive and receptive language function, and a multimodal brain imaging protocol will be implemented that includes structural imaging (MRI), functional imaging (fMRI), perfusion imaging (arterial spin labeling, ASL) and diffusion tensor imaging (DTI). The proposed research has two specific aims. The first is to identify the neural changes that are associated with successful recovery from aphasia in the acute and subacute periods after a stroke. The second aim is to determine the initial imaging and behavioral variables that are most predictive of the extent of eventual recovery of language function. We hypothesize that the inclusion of fMRI, ASL and DTI in combination with comprehensive language assessments will provide a more complete characterization of the brain's changing state than standard clinical imaging and language measures, and will explain a greater proportion of the variance in recovery patterns. A better understanding of the neural correlates of successful recovery will improve the accuracy of prognoses so as to better plan medical treatments and behavioral interventions.